lllyasviel / lllyasviel/FramePack

Future feature request - Something like ControlNet

Open
#516 0 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
17.3k
Forks
1.7k
PR merge metrics
No merged PRs in 30d

Description

I don't know if this is even possible with these models, but I would love to see better face control - more consistent with source images. Its good, but I would love to see better. Something like a ControlNet for faces. Support for pose videos like dense pose or the stick figure pose videos, or video 2 video for movement control, would be great. But completely understand if its outside of the scope of this project or this model's capabilities. Might be able to get more consistent faces if you have a reference face that you reference every few sections. That should help with drift as the video gets longer.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No files, tests, or entry points are named. First define whether face-reference conditioning, dense pose or stick-figure pose videos, and video-to-video movement control are in scope; done would require an agreed, testable feature specification.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
computer-vision, machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
15/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.